OsPPR1, a pentatricopeptide repeat protein of rice is essentialfor the chloroplast biogenesis

Kodiveri M. Gothandam, Eun-Sook Kim, Hongjoo Cho and Yong-Yoon Chung*School of Life Sciences and Biotechnology, Korea University, Sungbuk-ku, 136-701, Seoul, Anam-Dong,Korea (*author for correspondence; e-mail ychung@korea.ac.kr)

Received 3 September 2004; accepted in revised form 18 April 2005

Key words: chloroplast mutant, pentatricopeptide repeat, rice, transit peptide

Abstract

In this paper, we report a novel pentatricopeptide repeat (PPR) protein gene in rice. PPR, a characteristicrepeat motif consisted of tandem 35 amino acids, has been found in various biological systems includingplant. Sequence analysis revealed that the gene designated OsPPR1 consisted of an open reading frame of2433 nucleotides encoding 810 amino acids that include 11 PPR motifs. Blast search result indicated thatthe gene did not align with any of the characterized PPR genes in plant. The OsPPR1 gene was found tocontain a putative chloroplast transit peptide in the N-terminal region, suggesting that the gene producttargets to the chloroplast. Southern blot hybridization indicated that the OsPPR1 is the member of a genefamily within the rice genome. Expression analysis and immunoblot analysis suggested that the OsPPR1was accumulated mainly in rice leaf. Antisense transgenic strategy was used to suppress the expression ofOsPPR1 and the resulted transgenic rice showed the typical phenotypes of chlorophyll-deficient mutants;albinism and lethality. Cytological observation using microscopy revealed that the antisense transgenicplant contained a significant defect in the chloroplast development. Taken together, the results suggest thatthe OsPPR1 is a nuclear gene of rice, encoding the PPR protein that might play a role in the chloroplastbiogenesis. This is the first report on the PPR protein required for the chloroplast biogenesis in rice.

Abbreviations: cTP, chloroplast transit peptide; ORF, open reading frame; PPR, pentatricopeptide repeat;RT-PCR, reverse-transcriptase polymerase chain reaction

Introduction

A new gene family called PPR was discovered bySmall and Peeters (2000) during a bioinformaticscreening of the Arabidopsis proteins. The pentat-ricopeptide repeat (PPR) family was named afterfinding the characteristic tandem 35-amino acidmotif that makes up the major part of each ofthese proteins. Although the PPR motif shows astructural similarity to the tetratricopeptide (TPR)motif, the bioinformatic approach has distin-guished PPR from the TPR (Fisk et al., 1999;Small and Peeters, 2000). The PPR motif is also

found in few animal and fungal proteins (Mantheyand McEwen, 1995; Coffin et al., 1997; Mantheyet al., 1998; Tsuchiya et al., 2002) but the familyhas greatly expanded in higher plants. The numberof PPR repeats per protein within the family variesfrom 2 to 26, and each PPR motif is predicted tocomprise two a-helices (Aubourg et al., 2000;Small and Peeters, 2000; Brown et al., 2003). ThePPR proteins have shown to influence stability ortranslation of specific organellar mRNAs (Smalland Peeters, 2000; Stern et al., 2004).

Currently, little is known about the functionsof PPR proteins, while over 450 genes encoding

Plant Molecular Biology (2005) 58:421–433 � Springer 2005DOI 10.1007/s11103-005-5702-5

PPR proteins are predicted in the Arabidopsis andrice genome (Stern et al., 2004). Majority of thePPR proteins are predicted to be targeted to eithermitochondria or chloroplasts. Studies of the PPRproteins that target to the mitochondria have beendemonstrated that the proteins are involved in therestoration of cytoplasmic male sterility. In petu-nia, the PPR-containing ORF, denotedRf-PPR592, in the Rf locus has shown to contain14 repeats of the PPR motif and to restore fertilityto CMS plants (Bentolila et al., 2002). It has alsobeen reported that, in rice, the Rf-1 gene encoding18 repeats of the PPR motif promotes the pro-cessing of aberrant atp6 RNA in a restorer line(Kazama and Toriyama, 2003). Several researchpapers have been published on the PPR proteinsthat function in the chloroplast. A report wasmade on the maize PPR protein, CRP1, that it isinvolved in the translation of the chloroplast petAand petD mRNAs and the processing of the petDmRNA from a polycistronic precursor (Fisk et al.,1999). Another PPR protein in maize, PPR2, wasshown to target to the chloroplast, regulatingaccumulation of the plastid ribosome (Williamsand Barkan, 2003). Other PPR proteins in Ara-bidopsis and radish were found to function in thechloroplast RNA processing (Lahmy et al., 2000;Meierhoff et al., 2003; Nakamura et al., 2003,2004). These results suggest that the PPR motifsare the nucleus-encoded factor that controls thechloroplast gene expression. PPR protein may alsocontrol the circadian rhythm in plant (Oguchiet al., 2004).

Our present study deals with the discovery of anovel PPR gene in rice. The encoded protein,OsPPR1, contains a chloroplast transit peptide atits N-terminal region. Experimental evidencesconfirm that the protein targets to the chloroplast.Antisense transgenic study shows that suppressionof the OsPPR1 expression causes a chlorophylldeficient mutant phenotype. Recently, two chlo-rophyll-deficient mutants have been reported inrice; knock-out of the OsCHLH gene expressionby T-DNA insertional mutagenesis and an anti-sense suppression of the OsHAP3 gene (Junget al., 2003; Miyoshi et al., 2003). Cytologicalanalysis using microscopy revealed that theOsPPR1 antisense transgenic plant contains sig-nificant defects in the chloroplast developmentthat differ from those found in the two chloro-phyll-deficient mutants. Our study suggests that

the OsPPR1 gene encoding the novel PPR proteinis essential for chloroplast biogenesis in rice.

Materials and methods

Sequence analysis of OsPPR1

The OsPPR1 nucleotide and amino acid sequenceswere analyzed by the Basic Local AlignmentSearch Tool (BLAST) and the Soft berry prog-rame (http://www.softberry.com/). The PPR re-gions were identified by Pfam programe (Batemanet al., 2002). The transit peptide and subcellularlocalization were predicted using the softwareprograme, ChloroP (Emanuelsson et al., 1999).The chloroplast transit peptide cleavage site waspredicted by the SignalP programe (Nielsen et al.,1997). Sequence comparisons were conductedusing Clustal W (Altschul et al., 1990).

Vector construction, transformation, and inversePCR analysis

A binary vector, pGA1611 (Kang et al., 1998), wasused for rice transformation. This vector is aderivative of the pGA482 (An et al., 1988) whichcontains the hygromycin phosphotransferase (hph)gene as a selectable marker under the control ofthe CaMV 35S promoter. The vector also containsseveral unique sites (HindIII, SacI, HpaI, andKpnI) between the maize ubiquitin promoter,including the first intron of the ubiquitin gene(Christensen et al., 1992), and the nopaline syn-thase (nos) terminator. The OsPPR1 cDNA clonewas inserted in between the SacI and KpnI sites ofpGA1611 in the antisense orientation. A japonicarice variety, Dongjin, was used for transformationby the Agrobacterium co-cultivation method (Jeonet al, 1999). Agrobacterium tumefaciens LBA4404,containing Ach5 chromosomal background and adisarmed helper-Ti plasmid pAL4404 was used forrice transformation (Hoekema et al., 1983). Calliwere induced from mature seeds on N6 medium.Agrobacterium-mediated transformation of ricecallus was performed according to the method ofLee et al. (1999). Transformed calli were selectedby hygromycin resistance, and the transgenicplants were regenerated from the transformedcalli. The T-DNA integration site in the genomeof the transgenic plant was located by an inverse

422

PCR analysis. To do this, 1 lg of genomic DNAwas digested with HindIII restriction endonuclease(see the Figure 4A). After ethanol precipitation,the DNA was then self-ligated by T4-DNA ligase.PCR reaction was carried out with 5 min dena-turation at 94 �C, followed by 35 cycles (eachcycle: 94 �C, 1 min; 62 �C, 1 min; and 72 �C,2 min), then a final extension for 10 min at 72 �C.The primers used in the inverse PCR reactionwere: 5¢-CTA GCT GAT AGT GAC CTT AGGCGA C-3¢ (P1) and 5¢-GTC GAT GCT CAC CCTGTT GTT TGG TG-3¢ (P2).

Construction of fusion protein and PEG-mediatedtransformation

The N terminal region about 780 bp of OsPPR1containing the chloroplast transit peptide wasfused in frame to the XbaI and BamHI restrictionenzyme sites of the N-terminus of green fluores-cent protein (GFP) in smGFP vector. The fusionconstructs were introduced into tobacco protop-lasts prepared from N. thaliana leaves by poly-ethylene glycol-mediated transformation (Jinet al., 2001). Expression of the fusion constructswas monitored using confocal laser scanningmicroscopy (Zeiss, Germany) based on the Bio-RAD MRC 600 confocal system. GFP wereexcited with the 488 nm laser line, and GFPemission was detected using a 505–530 nm band-pass filter. Additional filtering was obtained usinga 488 nm primary dichroic mirror, which reflectsthe laser light and transmitting fluorescence. AsmGFP vector without OsPPR1 was included asa control.

RT-PCR and genomic DNA blot analysis

Total RNA was isolated from rice flower, leaf,and root using TRIzol reagent. Five microgramsof the total RNA was used to synthesize single-stranded cDNA. RT-PCR was performed at94 �C incubation for 5 min followed by 30 cyclesof 94 �C for 30 s, 62 �C for 45 s, and 72 �C for50 s, and 5 min extension at 72 �C. Constitutivelyexpressed OsActin gene was used as normaliza-tion control. The OsPPR1 primers used were5¢-GAG GGA GTT CTG GAG AAT GCC AGTC-3¢ (forward) and 5¢-GAA TGC TTG CTGCCA CCA TAA GCT G-3¢ (reverse). Actinprimers were 5¢-TCC ATC TTG GCA TCT

CTC AG-3¢ (forward) and 5¢-GTA CCC GCATCA GGC ATC TG-3¢ (reverse). Primers for PsaA, 5¢-CAT GTG GAT TGG CGG ATT TC-3¢(forward) and 5¢-CAA CTT TGC CGC CTACTG CTA C-3¢ (reverse); for Psb B, 5¢-CTCATC ATA TTG CTG CGG GTA C-3¢ (forward)and 5¢-GTT CGC GCC CTT CTT TAT CTC-3¢(reverse); for Pet A, 5¢-GAG GGT TGA ATGTGG GTG CTG-3¢ (forward) and 5¢-CAC TTGGCG TCC ATC CGA TG-3¢ (reverse). GenomicDNA was isolated from rice seedlings as de-scribed by Dellaporta et al. (1983). Twentymicrograms of total genomic DNA was digestedwith EcoRI, HindIII, and XbaI restriction en-zymes, and the digested DNA fractions weresubjected to electrophoresis on 0.8% (w/v) aga-rose gel. The genomic DNA transfer was carriedout by the standard method (Sambrook et al.,1989) and the membrane was hybridized with 32P-labelled OsPPR1 cDNA probe at 60 �C (lowstringent condition) and 65 �C (high stringentcondition) according to Church and Gilbert(1984).

Immuno blot analysis

The total leaf protein were isolated from riceplants in the buffer containing 50 mM Tris–HCl,(pH 7.5), 1 mM EDTA, 8 mM MgCl2, and 1%b-Mercaptoethanol. Chloroplasts were isolated asdescribed by Yehudai-Resheff et al. (2001). Theconcentration of protein samples were determinedby the Bio-Rad DC Protein Assay. Proteinsamples were separated on 10% SDS-polyacryl-amide gel and transferred to polyvinylidenedifluoride (PVDF) membrane. The membranewas incubated with OsPPR1 rabbit polyclonalantiserum (diluted 1:10 000) for 2 h in TBSbuffer, containing 0.5% BSA [Primary antibodyagainst OsPPR1 was produced from rabbit byinjecting a synthetic oligopeptide derived fromthe OsPPR1 sequence (CNRSECLLKG-KEVHGHT)]. After two rinses in TBST(TBS + 0.5% Tween 20), the membrane wasincubated with anti-rabbit IgG-conjugated withhorseradish peroxidase (diluted 1:10 000) for 1 hin TBS buffer containing 0.5% BSA. After tworinses in the TBST, OsPPR1 was visualized onthe blot using the Alkaline Phosphate conjugatesubstrate kit (Bio-Rad). The same procedure wasfollowed for the aspartyl protease control.

423

Transmission electron microscopy

For electron microscopy, wild type and mutantrice leaves were fixed in a solution containing 2.5%glutaraldehyde, 4% para-formaldehyde in 0.1 Mphosphate buffer (pH 7.4) at 4 �C for 4 h. Theywere then rinsed in 0.1 M phosphate buffer (pH7.4) and further fixed in 1% (w/v) osmium tetrox-ide (OsO4) for 4 h at 4 �C. After rinsing in 0.1 Mphosphate buffer, the samples were dehydratedand embedded in LR white resin (London ResinCompany). Thin sections (40–50 nm thickness)were prepared with an ultra-microtome (LKB,Bromma 2088) and were collected on nickel grids(1-GN, 150 mesh). These sections were stainedwith uranyl acetate and lead citrate and examinedunder a transmission electron microscope (JEM-100CX-1).

Measurement of chlorophyll content

Chlorophyll was extracted from 2 weeks-grownseedling with 80% acetone. Absorbance of theextract was measured spectrophotometrically at645 and 663 nm. Chlorophyll contents were deter-mined according to Arnon (1949).

Results

Sequence analysis of OsPPR1

The OsPPR1 cDNA clone was isolated from a ricecDNA library. Sequence analysis of the OsPPR1clone revealed that the gene consists of 2870 bp ofthe nucleotides comprising the coding region of2,433 bp. The ORF of OsPPR1 encoded a poly-peptide of 810 amino acids with the predictedmolecular mass of 88.3 kD. Nucleotide anddeduced amino acid sequences of the OsPPR1was registered in the GenBank (accession numberAY584749). The Pfam for the functional domainanalysis (see the materials and methods) revealedthat OsPPR1 contains 11 PPR motifs. Databasesearches with the deduced OsPPR1 amino acidsshowed that the protein shares homology with alarge group of the PPR proteins from plants,including Arabidopsis and rice. Among them, asuccessful alignment was achieved only betweenthe OsPPR1 and an Arabidopsis hypotheticalprotein (At1g74600), suggesting the gene has not

been characterized yet in plant (Figure 1A). Wefound that in the OsPPR1, the PPR motifs exist ina varying degree of its conservation and that onlythe 1–2, 3–4 and 8–10 repeats were tandem(Figure 1B and D). The consensus motif derivedfrom the 11 PPRs of OsPPR1 was also found toshow a high degree of homology with that fromArabidopsis PPRs (Small and Peeters, 2000; Fig-ure 1C). The ChloroP analysis revealed that theOsPPR1 contains the chloroplast targetingsequence with a putative cleavage site at Ala27

(Figure 1A). The results suggested that theOsPPR1 encodes a PPR protein targeted tochloroplast in rice.

Genomic organization and expression of OsPPR1

DNA blot analysis with the rice genomic DNAwas done to investigate the organization of Os-PPR1 in the rice genome. Analysis of the restric-tion enzyme digestion sites in the 3 kb of thegenomic DNA region which contains the entiresequence of OsPPR1 revealed that its single exondoes not contain the digestion sites of EcoRI andHindIII whereas one XbaI site is existed(Figure 2A). Thus, these three digestion sites wereused in the preparation of a DNA blot. The DNAblot was then hybridized at two different (low andhigh) stringent conditions (see materials andmethods). The DNA blot hybridization at thelow stringent condition showed that two bandswith the EcoRI and HindIII digestions and threebands with the XbaI digestion were hybridized,suggesting that the OsPPR1 gene forms a genefamily in the rice genome (Figure 2B, left). Thehigh stringent condition of the hybridization,however, allowed only one band to appear onthe blot, indicating that the OsPPR1 gene exists asa single copy in the rice genome (Figure 2B, right).To elucidate expression pattern of the OsPPR1gene, total RNAs from flower, leaf, and root wereextracted. Single stranded cDNAs were then syn-thesized using 5 lg of the total RNA from eachsample and used in RT-PCR experiment as thetemplates. The result revealed that the OsPPR1transcript was accumulated mainly in the leaf anda very limited amount of the transcript wasdetected in the rest of the organs, suggesting thatthe gene mainly functions in the rice leaf(Figure 2C). We have also done the RT-PCRanalysis with the rice leaves at the different

424

developmental stages, young, immature, andmature stages (data not shown). We did notdetect, however, any difference in the transcriptaccumulation among the leaf stages, suggestingthat the gene is constitutively expressed during theplant growth without regulation.

Subcellular localization of OsPPR1

Sequence analysis of the OsPPR1 revealed that theN-terminus of OsPPR1 contains a chloroplast

transit peptide (Figure 1A). Therefore, to investi-gate whether the OsPPR1 indeed targets to thechloroplast, a GFP fusion protein with theN-terminal region of OsPPR1 was constructed(Figure 3A) and transformed into tobacco proto-plast by the PEG-mediated transformation meth-od. Subcellular localization of the OsPPR1:GFPwas then detected with a confocal microscope forthe green and chlorophyll fluorescence. Merging ofthe GFP and the chlorophyll fluorescence imagesconfirmed the expected localization of OsPPR1 in

Figure 1. Sequence analysis of OsPPR1. (A) Alignment of OsPPR1 with the predicted PPR protein from Arabidopsis database.

The predicted cleavage site of the chloroplast transit peptide is indicated by a vertical arrow in the N-terminal region. Amino acid

residues used for the primary antibody production is indicated by asterisks. Identical amino acid residues are shaded in black and

similar ones are grey-shaded. (B) Alignment of 11 PPR motifs within the OsPPR1: Residues identical to the consensus (shown in

the C) are shaded in black and similar residues are shaded in grey. The numbering above the alignment indicates the position with-

in each repeating unit. Helices A (residues 1–13) and B (residues 17–19) are predicted from a secondary structure prediction prog-

rame. (C) Comparison of the consensus motif derived from the OsPPR1 with those of the maize PPR2, Arabidopsis PGR3, radish

P67, maize HCF152, and the PPR consensus suggested by Small and Peeters (2000). The amino acids same as those of the

OsPPR1 consensus motif are shaded in black. D. comparison of the tandem repeats found in the PPR proteins: 1. OsPPR1, 2.

Arabidopsis PPR (At1g74600; NP_177599), 3. radish P67 (CAC01928), 4. Maize CRP1 (T01685). 5. maize PPR2 (AAP37977). The

PPR motifs are shown in black boxes.

425

the chloroplast (Figure 3B). We also performed animmunoblot experiment with the total proteinextracts from leaf and chloroplast. To do this, apolyclonal antibody was raised to recognize theOsPPR1 protein and we found that the antibodydetected a protein band of approximately 88 kD,demonstrating its reliability. The immunoblotshowed that as anticipated the OsPPR1 proteinwas enriched in the chloroplast extract withrespect to its concentration in the leaf extract(Figure 3C). Taken together, the results indicatedthat the OsPPR1 targets to the chloroplast in rice.

Characterization of the antisense transgenic plant

In order to study the function of OsPPR1 gene inrice, we constructed a binary vector containing anantisense orientation of the OsPPR1 cDNAlocated under the maize ubi promoter (Figure 4A).The ubi promoter has shown a strong activity incereals such as rice, maize, oats, wheat, and sugarcane (McElroy and Brettell, 1994). We trans-formed the construct into rice by Agrobacterium-mediated transformation method and producedthe transgenic plants regenerated from twentyindividual calli. Interestingly, four of the trans-genic plants displayed a mutant phenotype inwhich they fail to accumulate chlorophylls, becom-ing white and lethal mutants (Figure 4A). Thechlorophyll mutant lines were used to analyze theexpression of the endogenous OsPPR1 gene. The

result revealed the complete absence of OsPPR1transcript in the chlorophyll mutant lines (Fig-ure 4B). Among them, one chlorophyll mutant linewas selected for further detailed study. In that linewe did confirm the integration of the T-DNA inthe transgenic plant by performing an inverse PCRanalysis (Figure 5A). The result revealed that theT-DNA has been integrated into a chromosomalDNA region between the two ORFs located on thechromosome 12 without interrupting any of theplastid-related genes. This result indicated that themutant phenotype was not caused by an unwantedmutation from the T-DNA blocking of anyplastid-related gene. Expression analysis of theendogenous OsPPR1 gene also revealed that asuccessive suppression of the OsPPR1 was noticedin the chlorophyll mutant plant, whereas the threepositive controls of the chloroplast-related genetranscripts were still detectable (Figure 5B). Takentogether, these results suggested that the osppr1mutation was resulted from the antisense expres-sion of the OsPPR1 transcript. We thus measuredand compared the amount of chlorophyll in theleaves from the chlorophyll-deficient transgenicline and the wild-type plant (Table 1). The resultindicated that amount of the chlorophylls wassignificantly reduced in the transgenic plant. Thisresult suggested that the OsPPR1 might have arole in the chloroplast biosynthesis. We performedan immunoblot analysis to verify the expression ofOsPPR1 protein in the chlorophyll-deficient trans-

Figure 2. Southern blot and RT-PCR analysis of OsPPR1. (A) Schematic representation of OsPPR1 gene in the rice genome: Posi-

tion of the restriction enzyme sites is shown with the specific abbreviations [EcoRI (E), HindIII (H) and XbaI (Xb)]. The OsPPR1

gene has a single exon which is indicated by a black-closed box. (B) Southern blot hybridization of the OsPPR1. Rice genomic

DNA (20 lg) was digested with the restriction enzymes mentioned above and then was separated on a 0.8% agarose gel. After

transferred to a nylon membrane, the DNA blot was hybridized with 32P-labelled OsPPR1 cDNA probe at 60 �C (left) and 65 �C(right). (C) Transcript accumulation of the OsPPR1 gene. The OsActin was used as a control.

426

genic line. The result showed that the OsPPR1protein was not detectable in the transgenic plant,indicating that the OsPPR1 expression was signif-icantly reduced (Figure 5C).

Cytological analysis of the leaf

Characterization of the transgenic plant stronglysuggested that the plant contains a significantdefect in the chloroplast development. We thustook a cytological analysis of the leaf using themicroscopy. The light microscopic observationrevealed that the overall leaf structure remainsnormal as shown in the wild-type leaf. Theycontained the two epidermal cell layers (upperand lower) and vascular bundles, and did not showany significant change in the size or number ofmesophyll cells, suggesting that blocking of the

OsPPR1 gene expression did not affect the leafformation (Figure 6A). We noticed, observing at alight microscopic level however, that the meso-phyll cells of the transgenic plant showed abnor-mal shape of the chloroplasts in a reduced number,while the wild-type mesophyll cells contained well-developed chloroplasts (Figure 6A and B). Inorder to investigate the abnormal structure of thechloroplast, we further carried out the cytologicalanalysis using the transmission electron micros-copy (TEM). The TEM analysis revealed that themesophyll cells of the transgenic plant containedproplastid-like structures instead of the normalchloroplasts which contain thylakoid membranesand starch granules, suggesting that their chloro-plast was underdeveloped (Figure 6C). Other sub-cellular organelles such as mitochondria andnucleus were found as normal as shown in the

Figure 3. Subcellular localization of OsPPR1. (A) Schematic diagram of the fusion constructs used in the experiment. SmGFP,

soluble modified GFP. (B) In vivo targeting of OsPPR1:GFP into protoplasts. Tobacco protoplasts were transformed separately

with SmGFP and OsPPR1:GFP and then examined after 16h of the transformation. (a) SmGFP in the transformed protoplast.

(b) OsPPR1:GFP in the chlorophyll channel, (c) OsPPR1:GFP in green channel, and (d) overlap of GFP (green) and chlorophyll

(red). (C) Immunoblot analysis of OsPPR1. From the top, the immunoblot probed with the anti-OsPPR1 antibody and the

Coomassie-stained gel of the large subunit of Rubisco. Lanes contained the indicated quantities (microgram) of total leaf or chlo-

roplast extract from wild type plant.

427

wild-type control plant grown under the samegrowth condition, indicating that the transgenicphenotype is restricted to the chloroplast (Fig-ure 6C). We did not observed any formation of theinternal membranes in the proplastid-like struc-tures (Figure 6E, G, and H). Taken together, theresults suggested that the OsPPR1 is strictlyresponsible for the chloroplast development.

Discussion

The PPR gene family has been identified only afew years ago in planta (Small and Peeters, 2000).Most of the gene products are predicted to betargeted to chloroplasts or mitochondria, suggest-ing their essential roles in organelle biogenesis(Lurin et al., 2004) and yet little is known aboutthe function of the proteins. This study presents amolecular characterization of the PPR gene,OsPPR1, isolated from rice. In plant, the PPRmotifs consist of 35 amino acid consensus

Figure 4. Phenotype and expression analysis of the antisense

transgenic plants. (A) Phenotype of the antisense transgenic

(left & middle) and wild type (right) plants. (B) Transcript

accumulation of the endogenous OsPPR1 in antisense trans-

genic lines and wild type. OsActin was used as a control.

Figure 5. Integration, expression and western analysis of the

antisense transgenic plant. (A) Schematic representation of

the T-DNA in the Binary vector pGA1611 and the integra-

tion position of the T-DNA in the chromosome 12. RB and

LB, the right and left border of the T-DNA, respectively; P1

and P2, the primers used in the inverse-PCR (the HindIII (H)

enzyme site was used in the PCR reaction). The OsPPR1

transcript was subcloned in between the Ubiquitin (Ubi) pro-

moter and nos terminator (Tnos) in the antisense orientation.

Hph, hygromycin phosphotransferase. ORF 1 and ORF 2

denote the Calcineurin B-like protein (AK111505) and a

hypothetical protein (AK073772), respectively. (B) Transcript

accumulation of the endogenous OsPPR1 and the plastidre-

lated genes (Psa A, Psb B, and Pet A) used as the positive

controls. Psa A, a subunit of photosystem I; Psb B, a subunit

of photosystem II; Pet A, a subunit of the cytochrome b6f

complex. AS, the antisense transgenic plant; WT, wild type

plant. OsActin was used as a loading control. (C) Detection

of the OsPPR1 protein in the antisense transgenic and wild

type plants. From the top, the blot probed with anti-OsPPR1

antibody; Coomassie-stained gel of the large subunit of Rubi-

sco; the blot probed with antiaspartyl protease (XP_467818)

antibody as a non-plastidial loading control. Lanes contained

the indicated quantities (microgram) of total leaf protein from

the antisense-and wild-type plants.

428

Table 1. Chlorophyll content of the wild type and the antisense plants.

Chlorophyll a Chlorophyll b Total chlorophyll

Wild type 2.25 ± 0.43 1.2 ± 0.39 3.45

Antisense plant 0.11 ± 0.02 0.031 ± 0.01 0.141

Chlorophylls were extracted and quantified by the Method of Arnon (1949). Values are expressed in micrograms pigment per milligramfresh weight of the leaf tissue. The average of five replicates is shown.

Figure 6. Microscopic observations of the antisense transgenic (A, C, E, G, and H) and wild type (B, D and F) leaf sections. Light

microscopic observation of the antisense transgenic (A) and wild type (B) leaves. TEM observation of the antisense transgenic (C

and E) and wild type (D and F) mesophyll cells. Two more micrographs (G and H) showed the proplastid-like structure. We did

not observe any internal membrane or prolamellar body. Cp, Chloroplast; Mt, Mitochondria, UE, Upper Epidermis; LE, Lower

epidermis; CW, Cell Wall; N, Nucleus; MC, Mesophyll Cell; St, Stroma; BC, Bulliform Cell; VB, Vascular Bundle; Thy, Thyla-

koid; SG, Starch Grain; Scale bars = 1 lm.

429

sequences with a characteristic distribution ofsmall side-chain amino acids, containing hydro-phobic and hydrophilic regions (Small and Peeters,2000). The OsPPR1 contains 11 PPR motifs thatare not tandem. Different types of PPR motifshave been reported recently in Arabidopsisnamely, normal PPR (P) motif, PPR-like S motif(shorter than PPR motif) and PPR-like L motif(longer than PPR motif) (Lurin et al., 2004). TheOsPPR1 showed the presence of five PPR-like Smotifs among the classical motifs. The protein wasfound to contain a chloroplast transit peptide(cTP) in its N-terminal region, suggesting that theprotein is one of the PPRs targeted to chloroplast.Previously few PPR genes that contain cTP werereported, although no PPR gene with cTP wasreported yet in rice; CRP1 (Fisk et al., 1999),PPR2 (Williams and Barkan, 2003), P67 (Lahmyet al., 2000), HCF152 (Meierhoff et al., 2003;Nakamura et al., 2003), and PGR3 (Yamazakiet al., 2004). The OsPPR1 also lacks the specific C-terminal sequence as predicted in some of theArabidopsis PPR proteins (Lurin et al., 2004).Genomic organization of the OsPPR1 geneshowed only one exon for its coding sequence,supporting the report that 80% of the ArabidopsisPPR genes contain only one exon (Lurin et al.,2004). Our expression analysis suggested that theOsPPR1 gene is preferentially expressed in the leaf(Figure 2). Immunoblot with the total extractsfrom leaf and chloroplast revealed that theOsPPR1 is mainly present in the chloroplast(Figure 3C). Furthermore, in vivo transformationof the OsPPR1:GFP fusion protein into tobaccoprotoplast revealed its intracellular targeting to thechloroplast (Figure 3B). This is consistent with thelocalization of the radish P67 in chloroplast(Lahmy et al., 2000). It has been further demon-strated that the PPR proteins are localized in thechloroplast stroma: HCF152, an RNA bindingPPR protein of Arabidopsis, was found to bespecifically localized in the stroma (Meierhoffet al., 2003). PPR2, a maize PPR protein requiredfor plastid ribosome accumulation, was also local-ized in the stroma (Williams and Barkan, 2003). Inour transient expression of OsPPR1:GFP, theGFP signal was appeared as uniformly distributedin the chloroplast. This suggests that the OsPPR1is likely present in the stroma. Taken together, theresults suggest that the OsPPR1 is a PPR proteintranslocated into the chloroplast.

Loss-of-function approach using antisensetechnology was chosen to deduce the role ofOsPPR1 during the chloroplast development.Antisense technology is based on blocking theinformation flow from DNA via RNA to protein.This technology has been applied to crop plants tomodify a variety of traits, such as flower color(Van der Krol et al., 1988), fruit ripening (Smithet al., 2002), reduction in allergic substances(Tada et al., 1996), and generating male sterilityfor hybrid seed production (Van der Meer et al.,1992; Xu et al., 1995; Bhalla and Smith, 1998). Itwas also reported that the efficiency of antisensestrategy increases with the degree of sequencehomology (Kim et al., 2000). In our experiment,with the antisense construct of OsPPR1 transcriptunder the maize ubi promoter, we found that theOsPPR1 expression was successively blocked bythe Agrobacterium-mediated transformation andthe resulted transgenic plant displayed a typicalchlorophyll-deficient mutant phenotype (Fig-ure 4). Further characterization of the osppr1mutant (i.e. the antisense plant) revealed that theplant contained a significant reduction in thechlorophyll content and showed severe defects inthe chloroplast development. Recently, we havereported a similar type of the chlorophyll-deficientmutant in rice by the T-DNA insertional muta-genesis (Jung et al., 2003). In the study, knock-outof the gene denoted OsCHLH which encodes thelargest subunit of the Mg-chelatase resulted in thereduction of the chlorophyll content and thefailure of the thylakoid membrane formation.Another rice mutant that shows the reduction ofchlorophyll accumulation with defects in thechloroplast development was reported by Kuratagroup (Miyoshi et al., 2003). The mutant gener-ated from the antisense transformation of theOsHAP3 gene showed a pale-green phenotype dueto the partial reduction of the chlorophyll accu-mulation. They found, however, that the mutantwas able to form internal membrane structures.These two mutants were consistent with the osppr1mutant in which they had all nucleus-encodedproteins targeted to the chloroplast and their genemutation caused problems in the chlorophyllaccumulation and the chloroplast development.The mutants, however, differ from the osppr1mutant in which the osppr1 chloroplast did notform any of the internal membranes nor a clearprolamellar body. Thus, it seems that the mutant is

430

arrested at an early stage of the plastid differen-tiation. In higher plants, when seedlings aregerminated in darkness, proplastids, the precursorof chloroplast, develop into etioplasts containingthe prolamellar body. Upon exposure to light, theetioplasts get rapidly converted into chloroplasts.At this stage, the prolamellar body further devel-ops into lamellae, and the stacking of thylakoidmembrane which contains the light-harvestingcomplexes is initiated along the lamellae. OurTEM observation showed that the mesophyll cellsof the osppr1 mutant contain only the proplastid-like structure instead of normal chloroplast (Fig-ure 6). Taken together, the results suggest that theOsPPR1 might play a role very early in thechloroplast biogenesis.

It has been demonstrated that PPR proteinscontain RNA-binding property (Meierhoff et al.,2003; Nakamura et al., 2003) as initially suggestedby Small and Peeters (2000). In maize as the closestmodel system of rice, a chloroplast-localized PPRprotein, PPR2, was reported that the protein isrequired for plastid ribosome accumulation andnot for RNA metabolism (Williams and Barkan,2003). The maize PPR2 contained 11 tandem PPRrepeats with a chloroplast transit peptide at its N-terminus. The OsPPR1 also contained 11 repeatsof PPR, but the protein was not structurallyrelated to the maize PPR protein, because onlythree of the PPRs were tandem in the OsPPR1(Figure 1D). This suggests that the OsPPR1 maynot act as the maize PPR2 because of theirdifferent helical structures formed by the repeatdomains. We have also examined whether theOsPPR1 gene responds to environmental stresses,such as salt (50–200 mM NaCl), temperaturevariations (16, 23, 28 and 32 �C), wounding(2, 4, 8, 20 h), hormones (IAA, 2,4-D, andNAA), and continuous light or dark conditions(40 h). From all these treatments, we did not findany regulation of the gene expression in rice (datanot shown), supporting the idea that the OsPPR1functions only in the chloroplast biogenesis. Fur-thermore, our study with the antisense plantdemonstrated that the gene plays a role early inthe chloroplast development of rice. Possibly, theOsPPR1 gene is involved in the processing ofplastid RNAs required for the early event ofchloroplast biogenesis. Further genetic and bio-chemical studies of the protein will be required togain insight into its detailed function.

Acknowledgements

This research was supported by a grant fromPlant Signal Network Research Center funded bythe Ministry of Science and Technology andfrom Biogreen 21 program by Rural Develop-ment Administration, Republic of Korea. Thiswork was also supported in part by a universitygrant (2003) from Korea University.

References

Akagi, H., Nakamura, A., Yokozeki-Misono, Y., Inagaki, A.,Takahashi, H., Mori, K. and Fujimura, T. 2004. Positionalcloning of the rice Rf-1 gene, a restorer of BT-typecytoplasmic male sterility that encodes a mitochondria-targeting PPR protein. Theor. Appl. Genet. 108: 1449–1457.

Altschul, S.F., Gish, W., Miller, W., Myers, E.W. andLipman, D.J. 1990. Basic local alignment search tool.J. Mol. Biol. 215: 403–410.

An, G., Ebert, P.R., Mitra, A. and Ha, S.B. 1988. Binaryvector. In: S.B. Gelvin and R.A. Schilperoort (Eds.), PlantMolecular Biology Manual, A3nd ed, Kluwer, Dordrecht,Netherlands, pp. 1–19.

Arnon, D.I. 1949. Copper enzymes in isolated chloroplasts:polyphenoloxidase in Beta vulgaris. Plant Physiol. 24: 1–15.

Aubourg, S., Boudet, N., Kreis, M. and Lecharny, A. 2000. InArabidopsis thaliana, 1% of the genome codes for a novelprotein family unique to plants. Plant Mol. Biol. 42: 603–613.

Bateman, A., Birney, E., Cerruti, L., Durbin, R., Etwiller, L.,Eddy, S.R., Griffiths-Jones, S., Howe, K.L., Marshall, M.and Sonnhammer., E.L.L. 2002. The Pfam protein familiesdatabase. Nucleic Acids Res. 30: 276–280.

Bentolila, S., Alfonso, A.A. and Hanson, M.R. 2002. Apentatricopeptide repeat-containing gene restores fertility tocytoplasmic male-sterile plants. Proc Natl Acad. Sci. 99:10887–10892.

Bhalla, P.L. and Smith., N. 1998. Agrobacterium tumefaciens-mediated transformation of cauliflower, Brassica oleraceavar. botrytis. Mol. Breeding 4: 531–541.

Brown, G.G., Formanova, N., Jin, H., Wargachuk, R.,Dendy, C., Patil, P., Laforest, M., Zhang, J., Cheung,W.Y. and Landry, B.S. 2003. The radish Rfo restorer gene ofOgura cytoplasmic male sterility encodes a protein withmultiple pentatricopeptide repeats. Plant J. 35: 262–272.

Christensen, A.H., Sharrock, R.A. and Quail, P.H. 1992. Maizepolyubiquitin genes: structure, thermal perturbation ofexpression and transcript splicing, and promoter activityfollowing transfer to protoplasts by electroporation. PlantMol. Biol. 18: 675–689.

Church, G.M. and Gilbert, W. 1984. Genomic sequencing.Proc. Natl. Acad. Sci. 81: 1991–1995.

Coffin, J.W., Dhillon, R., Ritzel, R.G. and Nargang, F.E. 1997.The Neurospora crassa cya-5 nuclear gene encodes a proteinwith a region of homology to the Saccharomyces cerevisiaePET309 protein and is required in a post-transcriptional stepfor the expression of the mitochondrially encoded COXIprotein. Curr. Genet. 32: 273–280.

431

Dellaporta, S.L., Wood, J. and Hicks, J.B. 1983. A plant DNAminipreparation: version II. Plant Mol. Biol. Rep. 1: 19–21.

Deslorie, S., Gherbi, H. and Laloui, W. 2003. Identification ofthe fertility restoration locus, Rfo, in radish as a member ofthe pentatricopeptide-repeat protein family. EMBO Rep. 4:1–7.

Emanuelsson, O., Nielsen, H. and von Heijne, G. 1999.ChloroP, a neural network-based method for predictingchloroplast transit peptides and their cleavage sites. ProteinSci. 8: 978–984.

Fisk, D.G., Walker, M.B. and Barkan, A. 1999. Molecularcloning of the maize gene crp1 reveals similarity betweenregulators of mitochondrial and chloroplast gene expression.EMBO J. 18: 2621–2630.

Hoekema, A., Hirsch, P.R., Hooykaas, P.J.J. and Schilperoort,R.A. 1983. A binary vector strategy based on separation ofvir- and T-region of the Agrobacterium tumefaciensTi-plasmid. Nature 303: 179–181.

Jeon, J.S., Chung, Y.Y, Lee, S., Yi, G.H., Oh, B.G. and An,G. 1999. Isolation and characterization of an anther-specific gene, RA8, from rice (Oryza sativa). Plant Mol.Biol. 39: 35–44.

Jin, J.B., Kim, Y.A., Kim, S.J., Lee, S.H., Kim, D.H., Cheong,G.-W. and Hwang, I. 2001. A new dynamin-like protein,ADL6, is involved in trafficking from the trans-Golginetwork to the central vacuole in Arabidopsis. Plant Cell13: 1511–1525.

Jung, K.H., Hur, J., Ryu, C.H., Choi, Y., Chung, Y.Y., Miyao,A., Hirochika, H. and An, G. 2003. Characterization of arice chlorophyll-deficient mutant using the T-DNA gene-trapsystem. Plant Cell Physiol. 44: 463–472.

Kang, H.-G., Jeon, J.-S., Lee, S. and An, G. 1998. Identifica-tion of class B and class C floral organ identity genes fromrice plants. Plant Mol. Biol. 38: 1021–1029.

Kazama, T. and Toriyama, K. 2003. A pentatricopeptiderepeat-containing gene that promotes the processing ofaberrant atp6 RNA of cytoplasmic male-sterile rice. FEBSLett. 544: 99–102.

Kim, C., Jeong, D.H. and An, G. 2000. Molecular cloning andcharacterization of OsLRK1 encoding a putative receptor-like protein kinase from Oryza sativa. Plant Sci. 152: 17–26.

Koizuka, N., Imai, R., Fujimoto, H., Hayakawa, T., Kimura,Y., Kohno-Murase, J., Sakai Kawasaki, T. S. and Imamura,J. 2003. Genetic characterization of a pentatricopeptiderepeat protein gene, orf687, that restores fertility in thecytoplasmic male-sterile Kosena radish. Plant J. 34: 407–415.

Komori, T., Ohta, S., Murai, N., Takakura, Y., Kuraya, Y.,Suzuki, S., Hiei, Y., Imaseki, H. and Nitta, N. 2004. Map-based cloning of a fertility restorer gene, Rf-1, in rice (Oryzasativa L.). Plant J. 37: 315–325.

Lahmy, S., Barneche, F., Derancourt, J., Filipowicz, W.,Delseny, M. and Echeverria, M. 2000. A chloroplasticRNA-binding protein is a new member of the PPR family.FEBS Lett. 480: 255–260.

Lee, S., Jeon, J.S., Jung,K.H. andAn,G. 1999.Binary vectors forefficient transformation of rice. J. Plant Biol. 42: 310–316.

Lurin, C., Andres, C., Aubourg, S., Bellaoui, M., Bitton, F.,Bruyere, C., Caboche, M., Debast, C., Gualberto, J.,Hoffmann, B., Lecharny, A., Ret, M.L., Martin-Magniette,M.L., Mireau, H., Peeters, N., Renou, J.P., Szurek, B.,Taconnat, L. and Small, I. 2004. Genome-wide analysis ofArabidopsis pentatricopeptide repeat proteins reveals theiressential role in organelle biogenesis. Plant Cell 16: 2089–2103.

Manthey, G.M. and McEwen, J.E. 1995. The product of thenuclear gene PET309 is required for translation of maturemRNA and stability or production of intron-containingRNAs derived from the mitochondrial COX1 locus ofSaccharomyces cerevisiae. EMBO J. 14: 4031–4043.

Manthey, G.M., Przybyla-Zawislak, B.D. and McEwen, J.E.1998. The Saccharomyces cerevisiae Pet309 protein isembedded in the mitochondrial inner membrane. Eur.J. Biochem. 255: 156–161.

McElroy, D. and Brettell, R.I.S. 1994. Foreign gene expressionin transgenic cereals. Trend Biotechnol. 12: 62–68.

Meierhoff, K., Felder, S., Nalamura, T., Bechtold, N. andSchuster, G. 2003. HCF152, an Arabidopsis RNA bindingpentatricopeptide repeat protein involved in the processingof chloroplast psbB-psbT-psbH-petB-PetD RNAs. Plant Cell15: 1480–1495.

Miyoshi, K., Ito, Y., Serizwa, A. and Kurata, N. 2003.OsHAP3 genes regulate chloroplast biogenesis in rice. PlantJ. 3: 532–540.

Nakamura, T., Meierhoff, K., Westhoff, P. and Schuster, G.2003. RNA-binding properties of HCF152, an ArabidopsisPPR protein involved in the processing of chloroplast RNA.Eur. J. Biochem. 270: 4070–4081.

Nakamura, T., Schuster, G., Sugiura, M. and Sugita, M. 2004.Chloroplast RNA-binding and pentatricopeptide repeatproteins. Biochem Soc Trans. 32: 571–574.

Nielsen, H., Engelbrecht, J., Brunak, S. and Heijne, G.von1997. Identification of prokaryotic and eukaryotic signalpeptides and prediction of their cleavage sites. Prot. Eng. 10:1–6.

Oguchi, T., Sage-Ono, K., Kamada, H. and Ono, M. 2004.Genomic structure of a novel Arabidopsis clock-controlledgene AtC401, which encodes a pentatricopeptide repeatprotein. Gene 330: 29–37.

Pyke, K.A. 1999. Plastid division and development. Plant Cell11: 549–556.

Sambrook, J., Fritsch, E. and Maniatis, T. 1989. Molecularcloning: A laboratory manual., 2nd ed., Cold spring HarborLaboratory Press, Cold Spring Harbor, New York.

Small, I.D. and Peeters, N. 2000. The PPR motif – a TPR-related motif prevalent in plant organellar proteins. TrendsBiochem. Sci. 25: 46–47.

Stern, D.B., Hanson, M.R. and Barkan, A. 2004. Genetics andgenomics of chloroplast biogenesis: maize as a model system.Trends Plant Sci. 9: 293–301.

Tada, Y., Nakase, M., Adachi, T., Nakamura, R., Shimada, H.,Takahashi, M., Fujimura, T. and Matsuda, T. 1996.Reduction of 14–16 kDa allergenic proteins in transgenicrice plants by antisense gene. FEBS Lett. 391: 341–345.

Tsuchiya, N., Fukuda, H., Sugimura, T., Nagao, M. andNakagama, H. 2002. LRP130, a protein containing ninepentatricopeptide repeat motifs, interacts with a single-stranded cytosine-rich sequence of mouse hyper variableminisatellite Pc-1. Eur. J. Biochem. 269: 2927–2933.

Van der Meer, I.M., Stam, M.E., van Tunen, A.J., Mol, J.N.and Stuitje, A.R. 1992. Antisense inhibition of flavonoidbiosynthesis in Petunia anthers results in male sterility. PlantCell 4: 253–262.

Williams, P.M. and Barkan, A. 2003. A chloroplast-localizedPPR protein required for plastid ribosome accumulation.Plant J. 36: 675–686.

Xu, H., Knox, R.B., Taylor, P.E. and Singh, M.B. 1995. Bcp1,a gene required for male fertility in Arabidopsis. Proc. Natl.Acad. Sci. 92: 2106–2110.

432

Yamazaki, H., Tasaka, M. and Shikanai, T. 2004. PPR motifsof the nucleus-encoded factor, PGR3, function in theselective and distinct steps of chloroplast gene expressionin Arabidopsis. Plant J. 38: 152–163.

Yehudai-Resheff, S., Hirsh, M. and Schuster, G. 2001. Poly-nucleotide phosphorylase functions as both an exonucleaseand a poly(A) polymerase in spinach chloroplasts. Mol. Cell.Biol. 21: 5408–5416.

433